This article presents the MAGI software package for inference of dynamic systems. The focus of MAGI is where the dynamics are modeled by nonlinear ordinary differential equations with unknown parameters. While such models are widely used in science and engineering, the available experimental data for parameter estimation may be noisy and sparse. Furthermore, some system components may be entirely unobserved. MAGI solves this inference problem with the help of manifold-constrained Gaussian processes within a Bayesian statistical framework, whereas the unobserved components have posed a significant challenge for existing software. We use three realistic examples to illustrate the functionality of MAGI. The user may choose to use the package in any of the R, MATLAB, and Python environments.
翻译:本文介绍了MAGI用于推断动态系统的软件包。MAGI的焦点是动态由非线性普通差异方程式和未知参数建模的地方。虽然这些模型在科学和工程中广泛使用,但可用于参数估计的实验数据可能很吵和稀少。此外,有些系统组件可能完全没有观测到。MAGI在贝叶斯统计框架内的多限制高斯进程的帮助下解决了这一推论问题,而未观测的构件对现有软件提出了重大挑战。我们用三个现实的例子来说明MAGI的功能。用户可以选择在任何R、MATLAB和Python环境中使用该包。